The Continual Reassessment Method for Multiple Toxicity Grades: A Bayesian Model Selection Approach

Grade information has been considered in Yuan et al. (2007) wherein they proposed a Quasi-CRM method to incorporate the grade toxicity information in phase I trials. A potential problem with the Quasi-CRM model is that the choice of skeleton may dramatically vary the performance of the CRM model, which results in similar consequences for the Quasi-CRM model. In this paper, we propose a new model by utilizing bayesian model selection approach – Robust Quasi-CRM model – to tackle the above-mentioned pitfall with the Quasi-CRM model. The Robust Quasi-CRM model literally inherits the BMA-CRM model proposed by Yin and Yuan (2009) to consider a parallel of skeletons for Quasi-CRM. The superior performance of Robust Quasi-CRM model was demonstrated by extensive simulation studies. We conclude that the proposed method can be freely used in real practice.     

Retrospective analysis of sequential dose-finding designs

The continual reassessment method (CRM) is a dose-finding design using a dynamic sequential updating scheme. In common with other dynamic schemes the method estimates a current dose level corresponding to some target percentile for experimentation. The estimate is based on all included subjects. This continual reevaluation is made possible by the use of a simple model. As it stands, neither the CRM, nor any of the other dynamic schemes, allow for the correct estimation of some target percentile, based on retrospective data apart from the exceptional situation in which the simplified model exactly generates the observations. In this article we focus on the very specific issue of retrospective analysis of data generated by some arbitrary mechanism and subsequently analyzed via the continual reassessment method. We show how this can be done consistently. The proposed methodology is not restricted to that particular design and is applicable to any sequential updating scheme in which dose levels are associated with percentiles via model inversion.     

Sequential designs for phase I clinical trials with late-onset toxicities

Traditional designs for phase I clinical trials require each patient (or small group of patients) to be completely followed before the next patient or group is assigned. In situations such as when evaluating late-onset effects of radiation or toxicities from chemopreventive agents, this may result in trials of impractically long duration. We propose a new method, called the time-to-event continual reassessment method (TITE-CRM), that allows patients to be entered in a staggered fashion. It is an extension of the continual reassessment method (CRM; O'Quigley, Pepe, and Fisher, 1990, Biometrics 46, 33-48). We also note that this time-to-toxicity approach can be applied to extend other designs for studies of short-term toxicities. We prove that the recommended dose given by the TITE-CRM converges to the correct level under certain conditions. A simulation study shows our method's accuracy and safety are comparable with CRM's while the former takes a much shorter trial duration: a trial that would take up to 12 years to complete by the CRM could be reduced to 2-4 years by our method.     

The continual reassessment method for multiple toxicity grades: a Bayesian quasi-likelihood approach

We consider the case of phase I trials for treatment of cancer or other severe diseases in which grade information is available about the severity of toxicity. Most dose allocation procedures dichotomize toxicity grades based on being dose limiting, which may not work well for severe and possibly irreversible toxicities such as renal, liver, and neurological toxicities, or toxicities with long duration. We propose a simple extension to the continual reassessment method (CRM), called the Quasi-CRM, to incorporate grade information. Toxicity grades are first converted to numeric scores that reflect their impacts on the dose allocation procedure, and then incorporated into the CRM using the quasi-Bernoulli likelihood. A simulation study demonstrates that the Quasi-CRM is superior to the standard CRM and comparable to a univariate version of the Bekele and Thall method (2004, Journal of the American Statistical Association 99, 26-35). We also present sensitivity analysis of the new method with respect to toxicity scores, and discuss practical issues such as extending the simple algorithmic up-and-down designs.     

Collagen I Self-Assembly: Revealing the Developing Structures that Generate Turbidity

Type I collagen gels are routinely used in biophysical studies and bioengineering applications. The structural and mechanical properties of these fibrillar matrices depend on the conditions under which collagen fibrillogenesis proceeds, and developing a fuller understanding of this process will enhance control over gel properties. Turbidity measurements have long been the method of choice for monitoring developing gels, whereas imaging methods are regularly used to visualize fully developed gels. In this study, turbidity and confocal reflectance microscopy (CRM) were simultaneously employed to track collagen fibrillogenesis and reconcile the information reported by the two techniques, with confocal fluorescence microscopy (CFM) used to supplement information about early events in fibrillogenesis. Time-lapse images of 0.5 mg/ml, 1.0 mg/ml, and 2.0 mg/ml acid-solubilized collagen I gels forming at 27°C, 32°C, and 37°C were collected. It was found that in situ turbidity measured in a scanning transmittance configuration was interchangeable with traditional turbidity measurements using a spectrophotometer. CRM and CFM were employed to reveal the structures responsible for the turbidity that develops during collagen self-assembly. Information from CRM and transmittance images was collapsed into straightforward single variables; total intensity in CRM images tracked turbidity development closely for all collagen gels investigated, and the two techniques were similarly sensitive to fibril number and dimension. Complementary CRM, CFM, and in situ turbidity measurements revealed that fibril and network formation occurred before substantial turbidity was present, and the majority of increasing turbidity during collagen self-assembly was due to increasing fibril thickness.     

Poly(A) polymerase phosphorylation is dependent on novel interactions with cyclins

We have previously shown that poly(A) polymerase (PAP) is negatively regulated by cyclin B-cdc2 kinase hyperphosphorylation in the M phase of the cell cycle. Here we show that cyclin B(1) binds PAP directly, and we demonstrate further that this interaction is mediated by a stretch of amino acids in PAP with homology to the cyclin recognition motif (CRM), a sequence previously shown in several cell cycle regulators to target specifically G(1)-phase-type cyclins. We find that PAP interacts with not only G(1)- but also G(2)-type cyclins via the CRM and is a substrate for phosphorylation by both types of cyclin-cdk pairs. PAP's CRM shows novel, concentration-dependent effects when introduced as an 8-mer peptide into binding and kinase assays. While higher concentrations of PAP's CRM block PAP-cyclin binding and phosphorylation, lower concentrations induce dramatic stimulation of both activities. Our data not only support the notion that PAP is directly regulated by cyclin-dependent kinases throughout the cell cycle but also introduce a novel type of CRM that functionally interacts with both G(1)- and G(2)-type cyclins in an unexpected way.     

Glutathione-S-transferase pi as a model protein for the characterisation of chemically reactive metabolites

Chemically reactive metabolites (CRMs) are thought to be responsible for a number of adverse drug reactions through modification of critical proteins. Methods that defined the chemistry of protein modification at an early stage would provide invaluable tools for drug safety assessment. Here, human GST pi (GSTP) was exploited as a model target protein to determine the chemical, biochemical and functional consequences of exposure to the hepatotoxic CRM of paracetamol (APAP), N-acetyl-p-benzoquinoneimine (NAPQI). Site-specific, dose-dependent modification of Cys47 in native and His-tagged GSTP was revealed by MS, and correlated with inhibition of glutathione (GSH) conjugating activity. In addition, the adaptation of iTRAQ labelling technology to define precisely the quantitative relationship between covalent modification and protein function is described. Multiple reaction monitoring (MRM)-MS of GSTP allowed high sensitivity detection of modified peptides at physiological levels of exposure. Finally, a bioengineered mutant cytochrome P450 with a broad spectrum of substrate specificities was used in an in vitro reaction system to bioactivate APAP: in this model, GSTP trapped the CRM and exhibited both reduced enzyme activity and site-specific modification of the protein. These studies provide the foundation for the development of novel test systems to predict the toxicological potential of CRMs produced by new therapeutic agents.     

Diphtheria toxin- and Pseudomonas A toxin-mediated apoptosis. ADP ribosylation of elongation factor-2 is required for DNA fragmentation and cell lysis and synergy with tumor necrosis factor-alpha.

We have reported that diphtheria toxin (DTX) mediates target cell lysis and intranucleosomal DNA fragmentation (apoptosis) and also synergizes with TNF-alpha. In this paper, we examined which step in the pathway of DTX-mediated inhibition of protein synthesis was important for induction of cytolytic activity and for synergy. Using a DTX-sensitive tumor cell line, we first examined the activity of the mutant CRM 197, which does not catalyze the ADP ribosylation of elongation factor-2 (EF-2). CRM 197 was not cytolytic for target cells and did not mediate intranucleosomal DNA fragmentation of viable cells. The failure of CRM 197 to mediate target cell lysis suggested that the catalytic activity of DTX is prerequisite for target cell lysis. This was corroborated by demonstrating that MeSAdo, which blocks the biosynthesis of diphthamide, inhibited DTX-mediated protein synthesis inhibition and also blocked target cell lysis. Furthermore, the addition of nicotinamide, which competes with NAD+ on the DTX action site of EF-2, also blocked DTX-mediated lysis. These findings suggest that ADP-ribosylation of EF-2 may be a necessary step in the pathway leading to target cell lysis. In contrast to the sensitive line, the SKOV-3 tumor cell line is sensitive to protein synthesis inhibition by DTX but is not susceptible to cytolysis and apoptosis by DTX. Thus, protein synthesis inhibition by DTX is not sufficient to mediate target cell lysis. The synergy in cytotoxicity obtained with the combination of DTX and TNF-alpha was examined in order to determine the pathway mediated by DTX in synergy. Like the direct lysis by DTX, synergy was significantly reduced by MeSAdo and by nicotinamide. Furthermore, synergy was not observed with combination of CRM 197 and TNF-alpha. These results demonstrate that, in synergy, DTX may utilize the same pathway required for its cytolytic activity. Pseudomonas aeruginosa exotoxin shared most the properties shown for DTX. Altogether, these findings demonstrate that DTX-mediated apoptosis is initiated at a step beyond the ADP ribosylation of EF-2.     

Overexpression and purification of the recombinant diphtheria toxin variant CRM197 in Escherichia coli

The expression of the recombinant diphtheria toxin mutant CRM197 in bacteria other than Corynebacterium diphtheriae has proven to be difficult. Here we propose a new and alternative procedure for the production of full-length CRM197 in Escherichia coli. The present study relates specifically to the expression of an artificial sequence and to a method for the isolation and purification of the corresponding protein. In particular, a synthetic gene coding for CRM197, bearing a short histidine tag and optimized for E. coli codon usage, was cloned in the pET9a vector. Accordingly, the over-expression of the protein was simply induced with arabinose in E. coli BL21AI. The recombinant protein was insoluble and always found inside protein aggregates, which were solubilised using urea. Surprisingly, the expression of CRM197, devoid of the short tag, always failed. Following a refolding step, the his-tagged CRM197 was purified by affinity and gel-filtration chromatography and the purity of the final preparation reached 95%. Interestingly, the recombinant protein features DNase activity, indicating that the presence of the tag is not affecting its biochemical properties. However, the removal of the synthetic tag could be easily obtained by incubating the target protein with a proper quantity of a commercial enterokinase.     

Single mutation in the A domain of diphtheria toxin results in a protein with altered membrane insertion behavior

The insertion of the A domain of diphtheria toxin into model membranes has been shown to be both pH- and temperature-dependent (Hu and Holmes (1984) J. Biol. Chem. 259, 12226-12233). In this report, the insertion behavior of two mutant proteins of diphtheria toxin, CRM197 and CRM9, was studied and compared to that of wild-type toxin. Results indicated that both CRM197 and CRM9 resembled toxin with respect to the pH-dependence of binding to negatively-charged liposomes at room temperature. However, CRM197 differed from toxin with respect to both the pH- and temperature-dependence of fragment A insertion; fragment A197 inserts more readily into the bilayer at 0 degrees C and low pH or at neutral pH and room temperature than does wild type fragment A under these same conditions. This result indicates that the single amino acid substitution in the A domain of CRM197 facilitates entry of fragment A197 into the membrane, suggesting that CRM197 may be conformationally distinct from native toxin. In fact, the fluorescence spectra of CRM197 and wild-type toxin as well as their respective tryptic peptide patterns indicate that, at pH 7, CRM197 more closely resembles the acid form of wild-type toxin than the native form of toxin. These data suggest that CRM197 may be naturally in a more 'insertion-competent' conformation. In contrast, the mutation in the B domain of CRM9 which results in a 1000-fold decrease in binding affinity for plasma membrane receptors apparently does not cause a change in either the insertion of fragment A9 or the lipid-binding properties of CRM9 relative to toxin.     

Continual reassessment designs with early termination

The continual reassessment method (CRM) is an increasingly popular approach for estimating the maximum tolerated dose (MTD) in phase I dose finding studies. In its original formulation, the scheme is based on a fixed sample size. Many experimenters feel that, whenever possible, it may be advantageous to bring these trials to an early halt and thus reduce average sample size required to complete the study. To address this issue a stopping rule has been proposed (O'Quigley and Reiner, 1998) based on the idea that continuing the study would not lead to a change in recommendation with high probability. The rule, based on precise probabilistic calculation, is quite involved and not straightforward to implement. A much simpler rule can be constructed based on the idea of having settled at some level. In this work we investigate more deeply the essential ingredients behind these rules and consider more closely their operating characteristics.     

The anti-tumor effect of cross-reacting material 197, an inhibitor of heparin-binding EGF-like growth factor, in human resistant ovarian cancer

Heparin-binding epidermal growth factor-like growth factor (HB-EGF) is a promising target for ovarian cancer therapy. Cross-reacting material 197 (CRM197), a specific HB-EGF inhibitor, has been proven to represent possible chemotherapeutic agent for ovarian cancer. However, the effect of CRM197 on the resistant ovarian carcinoma cells has not been sufficiently elucidated. Here, we found that HB-EGF was over-expressed in a paclitaxel-resistant human ovarian carcinoma cell line (A2780/Taxol) and a cisplatin-resistant cell line (A2780/CDDP), as well as the xenograft mouse tissue samples with these cells. To investigate the possible significance of the HB-EGF over-expression in A2780/Taxol and A2780/CDDP cells, we inhibited HB-EGF expression by CRM197 to investigate the effect of CRM197 treatment on these cells. We observed that CRM197 significantly induced anti-proliferative activity in a dose-dependent manner with the cell-cycle arrest at the G0/G1 phase and enhanced apoptosis in A2780/Taxol and A2780/CDDP cells. The sensitive ovarian carcinoma parental cell line (A2780), A2780/Taxol and A2780/CDDP cells formed tumors in nude mice, and enhanced tumorigenicity was observed in drug-resistant tumors. Furthermore, we observed that CRM197 significantly suppressed the growth of drug-resistant ovarian cancer xenografts in vivo (p<0.001). These results suggest that CRM197 as an HB-EGF-targeted agent has potent anti-tumor activity in paclitaxel- and cisplatin-resistant ovarian cancer which over-express HB-EGF.     

Overexpression and purification of the recombinant diphtheria toxin variant CRM197 in Escherichia coli

The expression of the recombinant diphtheria toxin mutant CRM197 in bacteria other than Corynebacterium diphtheriae has proven to be difficult. Here we propose a new and alternative procedure for the production of full-length CRM197 in Escherichia coli. The present study relates specifically to the expression of an artificial sequence and to a method for the isolation and purification of the corresponding protein. In particular, a synthetic gene coding for CRM197, bearing a short histidine tag and optimized for E. coli codon usage, was cloned in the pET9a vector. Accordingly, the over-expression of the protein was simply induced with arabinose in E. coli BL21AI. The recombinant protein was insoluble and always found inside protein aggregates, which were solubilised using urea. Surprisingly, the expression of CRM197, devoid of the short tag, always failed. Following a refolding step, the his-tagged CRM197 was purified by affinity and gel-filtration chromatography and the purity of the final preparation reached 95%. Interestingly, the recombinant protein features DNase activity, indicating that the presence of the tag is not affecting its biochemical properties. However, the removal of the synthetic tag could be easily obtained by incubating the target protein with a proper quantity of a commercial enterokinase.     

CRM1/Ran-mediated nuclear export of p27(Kip1) involves a nuclear export signal and links p27 export and proteolysis

We show that p27 localization is cell cycle regulated and we suggest that active CRM1/RanGTP-mediated nuclear export of p27 may be linked to cytoplasmic p27 proteolysis in early G1. p27 is nuclear in G0 and early G1 and appears transiently in the cytoplasm at the G1/S transition. Association of p27 with the exportin CRM1 was minimal in G0 and increased markedly during G1-to-S phase progression. Proteasome inhibition in mid-G1 did not impair nuclear import of p27, but led to accumulation of p27 in the cytoplasm, suggesting that export precedes degradation for at least part of the cellular p27 pool. p27-CRM1 binding and nuclear export were inhibited by S10A mutation but not by T187A mutation. A putative nuclear export sequence in p27 is identified whose mutation reduced p27-CRM1 interaction, nuclear export, and p27 degradation. Leptomycin B (LMB) did not inhibit p27-CRM1 binding, nor did it prevent p27 export in vitro or in heterokaryon assays. Prebinding of CRM1 to the HIV-1 Rev nuclear export sequence did not inhibit p27-CRM1 interaction, suggesting that p27 binds CRM1 at a non-LMB-sensitive motif. LMB increased total cellular p27 and may do so indirectly, through effects on other p27 regulatory proteins. These data suggest a model in which p27 undergoes active, CRM1-dependent nuclear export and cytoplasmic degradation in early G1. This would permit the incremental activation of cyclin E-Cdk2 leading to cyclin E-Cdk2-mediated T187 phosphorylation and p27 proteolysis in late G1 and S phase.     

Enhancer evolution in chordates: Lessons from functional analyses of cephalochordate cis-regulatory modules

Chordates comprise three major groups, cephalochordates (amphioxus), tunicates (urochordates), and vertebrates. Since cephalochordates were the early branching group, comparisons between amphioxus and other chordates help us to speculate about ancestral chordates. Here, I summarize accumulating data from functional studies analyzing amphioxus cis-regulatory modules (CRMs) in model systems of other chordate groups, such as mice, chickens, clawed frogs, fish, and ascidians. Conservatism and variability of CRM functions illustrate how gene regulatory networks have evolved in chordates. Amphioxus CRMs, which correspond to CRMs deeply conserved among animal phyla, govern reporter gene expression in conserved expression domains of the putative target gene in host animals. In addition, some CRMs located in similar genomic regions (intron, upstream, or downstream) also possess conserved activity, even though their sequences are divergent. These conservative CRM functions imply ancestral genomic structures and gene regulatory networks in chordates. However, interestingly, if expression patterns of amphioxus genes do not correspond to those of orthologs of experimental models, some amphioxus CRMs recapitulate expression patterns of amphioxus genes, but not those of endogenous genes, suggesting that these amphioxus CRMs are close to the ancestral states of chordate CRMs, while vertebrates/tunicates innovated new CRMs to reconstruct gene regulatory networks subsequent to the divergence of the cephalochordates. Alternatively, amphioxus CRMs may have secondarily lost ancestral CRM activity and evolved independently. These data help to solve fundamental questions of chordate evolution, such as neural crest cells, placodes, a forebrain/midbrain, and genome duplication. Experimental validation is crucial to verify CRM functions and evolution.     

Efficiency of New Dose Escalation Designs in Dose-Finding Phase I Trials of Molecularly Targeted Agents

Background

Statistical simulations have consistently demonstrated that new dose-escalation designs such as accelerated titration design (ATD) and continual reassessment method (CRM)-type designs outperform the standard “3+3” design in phase I cancer clinical trials.

Methods

We evaluated the actual efficiency of different dose escalation methods employed in first-in-human phase I clinical trials of targeted agents administered as single agents published over the last decade.

Results

Forty-nine per cent of the 84 retrieved trials used the standard “3+3” design. Newer designs used included ATD in 42%, modified CRM [mCRM] in 7%, and pharmacologically guided dose escalation in 1%. The median numbers of dose levels explored in trials using “3+3”, ATD and mCRM designs were 6, 8 and 10, respectively. More strikingly, the mean MTD to starting dose ratio appeared to be at least twice as high for trials using mCRM or ATD designs as for trials using a standard “3+3” design. Despite this, the mean number of patients exposed to a dose below the MTD was similar in trials using “3+3”, ATD and mCRM designs.

Conclusion

Our results support a more extensive implementation of innovative dose escalation designs such as mCRM and ATD in phase I cancer clinical trials of molecularly targeted agents.     

A curve-free method for phase I clinical trials

Consider the problem of finding the dose that is as high as possible subject to having a controlled rate of toxicity. The problem is commonplace in oncology Phase I clinical trials. Such a dose is often called the maximum tolerated dose (MTD) since it represents a necessary trade-off between efficacy and toxicity. The continual reassessment method (CRM) is an improvement over traditional up-and-down schemes for estimating the MTD. It is based on a Bayesian approach and on the assumption that the dose-toxicity relationship follows a specific response curve, e.g., the logistic or power curve. The purpose of this paper is to illustrate how the assumption of a specific curve used in the CRM is not necessary and can actually hinder the efficient use of prior inputs. An alternative curve-free method in which the probabilities of toxicity are modeled directly as an unknown multidimensional parameter is presented. To that purpose, a product-of-beta prior (PBP) is introduced and shown to bring about logical improvements. Practical improvements are illustrated by simulation results.     

CRM1 Is a Cellular Target of Curcumin: New Insights for the Myriad of Biological Effects of an Ancient Spice

Curcumin is the major constituent of turmeric plant, an ancient spice widely used in Indian cuisine and traditional herbal medicine. Recently, the potential medical use of curcumin as anti‐cancer and anti‐inflammatory agent has set off an upsurge in research into the mechanism for its broad biological effects. We showed that CRM1, an important nuclear exportin, is a cellular target of curcumin by serious experimental and theoretical investigation. Using a nuclear export functional assay, we observed a clear and rapid shift of cargo proteins from a cytoplasmic localization to the nucleus when treated with curcumin or its structural analogue dibenzylideneacetone (DBA). We demonstrated that curcumin could specifically target the conserved Cys⁵²⁸ of CRM1 through mass spectrometric analysis and in vivo experiments. Furthermore, computational modeling has revealed that curcumin could be correctly docked into the hydrophobic pocket of CRM1 judged from shape complementarity and putative molecular interactions. The Michael acceptor moiety on curcumin is within the appropriate distance to enable Michael reaction with Cys residue of CRM1. More importantly, we showed that nuclear retention of FOXO1 could be observed in the presence of Leptomycin B (LMB) or curcumin whereas in cells expressing the CRM1‐Cys⁵²⁸ mutant, only a cytoplasmic localization was observed. The inhibition of nuclear traffic by curcumin may account for its myriad of biological effects, particularly for its therapeutic properties in cancer and inflammatory diseases. Our findings may have important implications for further clinical investigation of curcumin .     

Quantitative analysis of protein synthesis inhibition and recovery in CRM107 immunotoxin-treated HeLa cells

Previously a mathematical model was proposed that quantitatively related protein synthesis inhibition kinetics of antitransferrin receptor-gelonin immunotoxins to the cellular trafficking of the targeting agent. That work is here extended to describe protein synthesis inhibition kinetics of immunotoxins containing the diphtheria toxin mutant CRM107. CRM107 differs from gelonin in both translocation and ribosomal inactivation mechanisms. Targeting agents used were antitransferrin monoclonal antibodies 5E9 and OKT9, OKT9Fab, and transferrin. CRM107 conjugates inhibited protein synthesis at substantially lower concentrations than gelonin conjugates; this effect was attributed to substantially higher translocation rates for CRM107. However, under certain conditions, CRM107 immunotoxin-treated cells were able to recover completely; this behavior was never observed with gelonin immunotoxins. To quantitatively capture this phenomenon, extracellular and cytosolic degradation of the toxin as well as growth-related recovery from toxin-induced damage were incorporated into the mathematical model. Translocation and cytosolic degradation rate constants were determined for each immunotoxin. Unlike the gelonin conjugates, the translocation rate of CRM107 conjugates depended on the targeting molecule. This provided indirect evidence that CRM107 remains disulfide linked to the targeting agent for at least part of the translocation process. Although the CRM107 conjugates all had higher translocation rates and inhibited protein synthesis at lower concentrations than the gelonin conjugates, the cells' ability to recover from protein synthesis inhibition at low immunotoxin concentrations limits the utility of CRM107 conjugates for targeted cell killing.